IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 54, NO.
4, JULY/AUGUST 2018
Standalone Photovoltaic Water Pumping System Using
Induction Motor Drive With Reduced Sensors
Bhim Singh , Fellow, IEEE, Utkarsh Sharma , Member, IEEE, and Shailendra Kumar , Member, IEEE
Abstract—A simple and efficient solar photovoltaic (PV) wa-
ter pumping system utilizing an induction motor drive (IMD) is
presented in this paper. This solar PV water pumping system com-
prises two stages of power conversion. The first stage extracts the
maximum power from a solar PV array by controlling the duty
ratio of a dc–dc boost converter. The dc bus voltage is maintained
by the controlling the motor speed. This regulation helps in the
reduction of motor losses by reducing motor currents at higher
voltage for the same power injection. To control the duty ratio, an
incremental conductance based maximum power point tracking
(MPPT) control technique is utilized. A scalar-controlled voltage
source inverter serves the purpose of operating an IMD. The stator
frequency reference of IMD is generated by the proposed control
scheme. The proposed system is modeled, and its performance is
simulated in detail. The scalar control eliminates the requirement
of a speed sensor/encoder. Consequently, the need of motor current
sensor is also eliminated. Moreover, the dynamics are improved by
an additional speed feedforward term in the control scheme. The
proposed control scheme makes the system inherently immune to
the variation in the pump constant. The prototype of PV-powered
IMD emulating the pump characteristics is developed in the labo-
ratory to examine the performance under different operating con-
ditions.
Index Terms—Induction motor drives, maximum power point
tracking (MPPT), photovoltaic (PV) cells, scalar control, water
pumping.
I. INTRODUCTION
HE RISING energy crises throughout the world and pol-
T lution of natural habitats have attracted considerable at-
tention from engineering and science fraternity since a couple
of decades. The knowledge for manifestation of renewable en-
ergy sources into a useful form has been developing rapidly.
The advent of fast switching power electronic devices and
the development in semiconductor technology have contributed
majorly to energy conversion methods. The renewable energy
Manuscript received July 6, 2017; revised September 19, 2017 and January 4,
2018; accepted March 12, 2018. Date of publication April 9, 2018; date of cur-
rent version July 17, 2018. Paper 2017-SECSC-0580.R2, presented at the 2016
IEEE 1st International Conference on Power Electronics, Intelligent Control
and Energy Systems, Delhi, India, Jul. 4–6, and approved for publication in the
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS by the Sustainable Energy
Conversion Systems Committee of the IEEE Industry Applications Society. This
work was supported by the Department of Science and Technology, Govern-
ment of India under Grant RP03128G, Grant RP02926, and Grant RP03222G.
(Corresponding author: Utkarsh Sharma.)
The authors are with the Department of Electrical Engineering, Indian
Institute of Technology Delhi, New Delhi 110016, India (e-mail:, bsingh@
ee.iitd.ac.in; utkarsh.kota@gmail.com; er.dwivedi88@gmail.com).
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TIA.2018.2825285
0093-9994 © 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications standards/publications/rights/index.html for more information.
Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
3645
utilization, which started from converting the energy of running
water, has traveled across to convert solar energy to electrical
energy directly today.
Earlier solar photovoltaic (PV) energy converters have been
inefficient, with efficiency as low as 5–6%, and highly expensive
[1]. However, with the advancement in technological research,
the efficiency of a PV array, at present, has reached 15–16%.
In addition, the prices have been reducing gradually. Today, PV
energy conversion is viewed as one of the promising alternatives
to fossil fuel based electricity generating systems, since there
are no toxic emissions, no greenhouse gases emission, no fuel
cost involvement, least maintenance cost, no water usage, and so
on. However, the technology is in developing phase and many
challenges need to be addressed, such as intermittency, high
initial cost, and low efficiency.
The solar water pumps [2]–[4] are gaining popularity in rural
areas, where the electricity is not available. Moreover, solar PV-
fed water pumps are favored in remote areas for irrigation, water
treatment plant, and agricultural purposes. For a country like
India, where 70% of the population depends upon agriculture,
irrigation is necessary for good yield. There are a large number
of water pumps in the world running with electricity or with
nonrenewable energy sources. The acquisitions of solar PV-
based water pumping systems [5] are more convenient compared
to diesel-based water pumping systems considering the cost and
pollution.
The design of a motor drive system powered directly from a
PV source demands creative solutions to face the challenge of
operation under variable power restrictions while maximizing
the energy produced and the amount of water pumped [6].
In PV pumping (PVP) systems, an induction motor drive
(IMD) shows good performance compared to other commer-
cial motors because of its rugged construction. The evolution
is intended to develop a productive, reliable, maintenance-free,
and cheap PV water pumping system [7]. However, new perma-
nent magnet motors such as brushless dc motors and permanent
magnet sine fed motors are used for pumping, but these are
still overshadowed by an induction motor because of cost and
availability constraints [8]. Moreover, the manufacturing of the
induction motor is in a mature stage giving an edge to its use in
developing countries for solar water pumping application. With
the emergence of outperforming solid state switches, high-speed
processors, and efficient motor control algorithms, IMD-based
water pumping systems have taken a step ahead to conventional
water pumping systems. Moreover, PV array fed IMD has per-
formed ruggedly in the field of pumping system by utilizing a
 3646
voltage source inverter (VSI). The proposed work deals with
a three-phase IMD for solar water pumping, which meets the
requirement of life without electricity in remote locations.
The initial cost of a solar power plant is high. Therefore,
once the plant is installed, the objective is to obtain the peak
power from the solar panels of the installed capacity. The de-
veloped water pumping system powered directly from the PV
array requires maximum power point tracking (MPPT) algo-
rithms to operate under different irradiation levels and to extract
the peak power from the solar PV array. Some of these MPPT
algorithms are recommended in [9]. A comparative study on
different MPPT techniques is provided in [10]–[12]. From the
operational point of view, MPPT is a mandatory segment of a
PV system. Substantial research is reported in past few years
in the area of MPPT. In this paper, an incremental conductance
(INC) based technique is used to obtain the peak power from the
solar PV array. Therefore, the proposed PV fed water pumping
system produces peak torque even at low radiation. The INC
technique is based on the comparison of output conductance of
the solar PV array to the INC. As compared to solar PV grid
interfaced systems [13], the major challenge in PV water pump-
ing is timely control of active power. This is due to the fact that
the mechanical time constant of the motor pump system is much
higher than that of the aforementioned system. Under sudden
fall in solar insolation, the PV array voltage tends to reduce
drastically, and consequently the level of flux in the motor falls
rapidly. Once the flux has decreased, the motor starts drawing
higher current, which is limited by the short-circuit current of
the PV array in order to rebuild the flux. The operating point
in the I–V curves of the PV array shifts to the current source
region as demonstrated by short-circuit current and very low
voltage. Due to insufficient power, the motor starts operating in
an unstable zone of torque–speed characteristics near to a point
where slip = 1. This particular condition is hazardous for the
motor’s health, and once the motor enters this zone, there has to
be a provision in the control, which can identify this condition
and restart the motor from the standstill condition. The motor
entering into such situations frequently would reduce the over-
all duty of the pump; hence, it is the responsibility of MPPT
algorithm to take care of such events.
To control the IMD-tied VSI, a simple V/f (voltage/frequency)
control approach is utilized in [14] and [15]. The pumping
system with a dc–dc converter and VSI is used for water
pumping application in [16]–[18]. However, the presented ap-
proach suffers from dc link voltage instability. The V/f approach
is simple, easy to implement, and cost effective. Dual invert-
ers are used to supply power to a centrifugal pump with the
sample-averaged zero-sequence elimination pulsewidth mod-
ulation technique [19]. Apart from V/f control, direct torque
control (DTC) and vector control techniques are complicated
and they require extra current sensors for implementation [20].
In V/f control, only PV array current, voltage, and dc bus volt-
age are sensed. The proposed system tracks the MPP point by
altering the modulation frequency so that the IMD is able to
extract the maximum power from the solar PV array at sus-
tained torque for different solar insolation levels. The proposed
system is able to supply more water compared to a solar PV
Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 54, NO. 4, JULY/AUGUST 2018
Fig. 1. System architecture for the standalone solar water pumping system.
fed dc motor based water pump. By utilizing V/f control, the
starting performance of the IMD is improved even if IMD is
started with lower solar insolation. Therefore, water is perma-
nently pumped from morning till evening. The starting current
of the induction motor connected to the fixed voltage ac mains
is around 5–6 times of full load current. Therefore, to start the
motor without any control, a higher number of solar modules
are required, whereas smooth start of the induction motor is
possible by using V/f control without drawing high starting cur-
rent. This also improves the life of the motor. Moreover, the
areas that are favored by the electrical connectivity may utilize
the grid-interfaced PVPs [21]. In Indian context, still many in-
door villages and agricultural lands do not possess a privilege
of having electrical network.
II. DESIGN OF THE PROPOSED SYSTEM
The system configuration of the PV water pumping system
is depicted in Fig. 1. It consists of a PV array followed by a
boost converter. A VSI is used to provide pulsewidth modulated
voltage input to the motor and pump assembly. The power from
the PV array is regulated using the INC method to attain its
maximum value with available radiation. The V/f control is used
to give reference speed to IMD.
A. Design of the Solar PV Array
An induction motor of 2.2 kW is selected for the proposed
system. If losses of the motor and pump are neglected, the
capacity of the PV array should be equivalent to the motor
capacity. In this case, a 2.4-kW PV array is selected
Pmp = (Np × Imp ) × (Ns × Vmp ) = 2.4 kW (1)
where Pmp is the maximum power that can be drawn from panels
at a given radiation, Vmp is the PV panel voltage at MPP, and Imp
is the current at MPP. Ns and Np are the number of modules
connected in series and parallel, respectively. Considering an
open-circuit voltage of the panel to be near to a dc link voltage
and the power drawn from the panel to be 2.4 kW, the number
of modules in series and parallel are selected to be 11 and 1,
respectively. The individual module and array specifications are
provided in Table I.
SINGH et al.: STANDALONE PHOTOVOLTAIC WATER PUMPING SYSTEM USING INDUCTION MOTOR DRIVE WITH REDUCED SENSORS 3647

TABLE I where TL is the load toque of water pump, which is equal to

SPECIFICATIONS OF THE SOLAR MODULE AND ARRAY
the torque offered by an induction motor under steady-state
operation and ωr is the rotational speed of the rotor in rad/s.
Since the rated torque and rated speed of the induction motor
are 14.69 N · m and 1430 rpm, respectively, the proportionality
constant Kpump is estimated using (6) as
14.69
Kpump = = 6.55 × 10−4 N · m/(rad/s)2 .
(2π1430/60)2
So, the proportionality constant is selected to be
6.55 × 10−4 N · m/(rad/s)2 .

B. Selection of DC Link Voltage III. CONTROL SCHEME FOR THE PROPOSED SYSTEM
The dc bus voltage of VSI is estimated from the following The proposed topology is a two-stage power conversion sys-
relation: tem for a solar PV array fed water pumping. It embodies scalar
Vdc V L -L control for IMD operation and the INC method for maximum
m× √ = √ (2) power extraction from the PV array. The simplicity and ease of
2 2 3 implementation of scalar control overshadow precise but com-
where m is the modulation index and VL -L is the line voltage putation intensive control algorithms such as vector control and
across the √motor terminals. Hence, direct torque control. Moreover, in the later algorithms, the posi-
Vdc = 2√32 × 230 = 375 V, which is the voltage required tion sensorless operation is itself an exhaustive task. The voltage
when modulation index is 1. The dc link voltage is chosen and current of PV array are sensed and fed to the INC algorithm.
to be 400 V. Based on the change in voltage, current, and power, this algo-
rithm decides the duty ratio of the boost converter. The boost
C. Design of DC Link Capacitor converter output voltage is maintained at a constant value using
The dc link capacitor is supposed to provide sufficient energy a proportional-integral (PI) controller. Since the pump charac-
at the time of transients such as fall in radiation and an increase teristics are centrifugal in nature, the power absorbed and the
in the load. Its value is calculated as [22] speed of the pump have a direct relation, as mentioned in (6).
A speed feedforward term is calculated from the available PV
1  
Cdc Vdc∗2 − Vdc1
2
= 3αV It power from which the PI controller output is subtracted. This
2 is helpful in reducing the burden on the PI controller and im-
1   proving the dynamic performance of the system. The V/f control
Cdc 4002 − 3752 = 3 × 1.2 × 133 × 8.2 × 0.005
2 algorithm generates the switching logic for VSI using sinusoidal
Cdc = 2026 μF. (3) pulsewidth modulation (PWM). If dc link voltage is higher than
the reference value, the PI controller increases the reference
In above expression, Vdc∗
refers to the set dc bus voltage, while speed given to V/f control and vice versa. The sum of two quan-
Vdc1 is the acceptable to the lower most voltage during tran- tities gives a resultant speed reference f ∗ for IMD, which is
sients. Morever, α is an overloading factor and t is the duration fed to the V/f control algorithm. The dc link voltage error is
of transient. estimated as

D. Selection of the DC–DC Boost Converter Vdcr = Vdc∗ − Vdc . (7)

The boost inductor duty cycle D is given as [23] The output of the dc link voltage PI controller is
 
Vdc − Vmp 400 − 373 ωdcr(n ) = ωdcr(n −1) + kp Vdcr (n ) − Vdcr(n −1) + ki Vdcr(n ) .
D= = = 0.0675 (4)
Vdc 400 (8)
Vmp D 372.9 × 0.0677 The speed term corresponding to PV power is
Lm = = = 1.875 mH. (5) 
ΔI1 fs (0.2 × 7.6 × 10 000)
ωP = K 3 PPV (9)
where fs is the switching frequency and ΔI1 is the amount of
ripple current. Thus, the inductance L value is selected to be where constant K is derived from the pump constant. The refer-
3 mH. ence frequency of the IMD is
1
E. Design of the Pump f∗ = (ωp − ωdcr ) . (10)
2π
For the selected water pump, the proportionality constant Initially, the boost converter pulses are kept OFF such that the
Kpump is given as system works as a single stage system and the speed is ramped
TL up to a threshold speed. After threshold speed, the control of
Kpump = (6) the boost converter is activated and the duty ratio is calculated
ωr 2

Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
 3648 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 54, NO. 4, JULY/AUGUST 2018

Fig. 2. P–V and I–V characteristics of the SPV array.

from the INC algorithm. This is realized to avoid high current

at starting since the MPPT algorithm gives maximum power
even at start. Using ramp frequency start, the starting current of
the motor is limited, which in the case of direct online starting
(DOL) at rated frequency is about 5–6 times that of the rated
current. Moreover, it prevents the solar PV array to go into Fig. 3. Flowchart for the INC algorithm for MPPT.
current source region at starting as the current drawn is very
high in DOL starting.
On the right side of MPP, slope is negative, which suggests
that dVdI PV
PV
< − VI PV
PV
, and on the left side, slope is positive, i.e.,
A. INC Method for MPPT dI PV
> − I PV
. At dI PV
MPP, slope is zero, i.e., dV = − VI PV . The
dV PV V PV PV PV
The solar PV array has nonlinear bell-shaped PPV versus VPV duty ratio of the boost converter is adjusted in accordance with
characteristics, as shown in Fig. 2. At any moment, the operating the algorithm, as shown in Fig. 3.
point depends on the impedance of the load connected to the
array terminals. A dc–dc converter is used to track the point of B. Scalar (V/f) Control of Induction Motor
operation on the PV curve. There have been many algorithms The scalar control of an induction motor is the most common
in the literature for tracking of maximum power point. Most and simplest control so far. Usually, induction motors are de-
basic of all is the perturb and observe algorithm, which involves signed for 50-Hz input voltage. For operation at a lower speed,
step change in the reference voltage or duty ratio to the dc–dc the voltage has to be reduced. The frequency control along with
converter and monitoring of the power output. It faces several voltage magnitude control is also desired for constant flux op-
issues when radiation changes. The INC method works much eration. The voltage should be proportional to the frequency
better in dynamic changes in solar insolation. This is due to a such that flux magnitude is maintained constant as ψs = V /ω.
fact mentioned in Section I that the mechanical time constant An IM is usually fed from a three-phase PWM VSI. Only an
of the motor is much higher than the electrical time constant of input parameter is the reference speed. Neglecting the small
the whole system. This work uses an INC algorithm, which is slip speed, the speed of the motor is approximately equal to the
based on the monitoring of slope of the PPV –VPV curve. reference speed. The speed reference is integrated to generate
On the right-hand side of the MPP, the slope is negative, while θ, which is used to obtain three sinusoidal voltage references,
on the left-hand side the slope is positive. At the point, where which are compared with high-frequency triangular wave to
maximum power is being transferred from the array, the slope generate the switching pulses for VSI. The speed reference is
of the curve is zero. With the change in radiation, at MPP, IPV estimated from the control scheme as mentioned in the previous
changes drastically, whereas VPV remains constant. Considering subsection.
a power equation and differentiating it with respect to voltage, 
relations between an INC and conductance are obtained for θ = ω ∗ dt. (14)
different sections of the curve as
Three phase reference voltages are
PPV = VPV × IPV (11)
Va∗ = m × sin (θ) (15)
dPPV dIPV
= IPV + VPV =0 (12) Vb∗ = m × sin (θ − 120 )◦
(16)
dVPV dVPV
dIPV IPV Vc∗ = m × sin (θ − 240◦ ) (17)
=− . (13) ∗
dVPV VPV where m = kf ω and m is the modulation index.

Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
SINGH et al.: STANDALONE PHOTOVOLTAIC WATER PUMPING SYSTEM USING INDUCTION MOTOR DRIVE WITH REDUCED SENSORS
Fig. 4. Starting performance of the proposed system.
IV. RESULTS AND DISCUSSION
Performance of a two-stage PV-fed water pumping system is
evaluated using the simulation package. The proposed system
is designed, modeled, and simulated in the MATLAB/Simulink
environment. The step change in the solar radiation is also sim-
ulated in order to determine the satisfactory performance of the
system under dynamic conditions.
A. Starting Performance of the Proposed System
Fig. 4 shows various parameters of the proposed water pump-
ing system at 500 W/m2 radiation. The dc link of VSI is ener-
gized initially. Since the switching device of the boost converter
is OFF, the voltage across the dc link of VSI is the open circuit
voltage of the PV array. It starts falling once the motor speed
increases. The PV array current starts from zero and rises up to
Imp . The PV voltage reaches Vmp once a threshold frequency is
passed, and the control of the boost converter is activated for
MPPT. At t = 8 s, the boost converter is activated and the sys-
tem reaches corresponding MPP. The dc link voltage is settled
at reference value because of the action of PI controller. It is
verified from the figure that the motor current never exceeds the
rated current, which is by the virtue of soft start. This practice
improves the lifespan of the motor.
B. Steady-State and Dynamic Performances of the
Proposed System
The behavior of the proposed standalone PV water pumping
system is shown in Fig. 5. This figure comprises simulation of
varied solar insolation changes. From t = 1 s to 2 s, the solar
insolation is constant at 800 W/m2 . The PV indices are at the
corresponding MPP. At t = 2 s, a slope decrement in the solar
insolation is simulated to test the MPPT algorithm effectiveness.
The PV voltage observes negligible change, while the PV cur-
rent varies proportional to the available insolation. Moreover,
the dc bus voltage is also maintained at reference voltage of
Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
3649
400 V without any failure. The speed and torque of the motor
are reduced with the reduction in PV power. This continues to
happen till t = 4 s, from where the system experiences a slope
increase in the solar insolation. Similar to the previous behavior,
the PV current starts increasing proportional to the solar radia-
tion, while there is not much change in the PV voltage. Conse-
quently, the available power from a PV source ramps up along
with the motor speed and motor torque. At t = 6 s, the system
operates in a steady state at a solar radiation of 1000 W/m2 . The
system faces a step decrement in the solar insolation from 1000
to 500 W/m2 at t = 7 s, owing to which the PV current reduces
instantly. However, still the PV voltage does not face many tran-
sients. The dc bus voltage experiences a slight transient change;
however, it restores to a reference voltage quickly. It is notewor-
thy that the dc bus voltage is maintained even at 50% reduction
in rated power. Similarly, a step increase in a solar insolation is
observed at t = 9 s. As anticipated from the previous behavior
of the system, the dc bus voltage is maintained to a reference
value, while there is no significant change in the PV voltage.
The motor speed and torque increase proportionally to balance
a power from a source.
V. EXPERIMENTAL VALIDATION OF THE PROPOSED SYSTEM
For the verification of the proposed configuration and the
control of proposed system, a prototype is developed in the
laboratory, which consists of a solar PV simulator (AMETEK
ETS 600 × 17DPVF), a boost converter, a VSI (Semikron),
a DSP (dSPACE 1104 real-time controller), and a three-phase
induction motor coupled to a dc generator.
A volumetric pump is realized by loading the three-phase
IMD using a dc generator. A resistive load is set to extract the
rated power from the PV source corresponding to the head of
pump. The solar PV array characteristics are designed in the
simulator software to provide a maximum power of 2.4 kW
with an open circuit voltage of 420 V and a short circuit of 7 A.
Hall-effect voltage and current sensors (LV-25P and LA-55P,
respectively) are used to sense the PV voltage, PV current, and
dc link voltage. An optocoupler-based isolation is provided be-
tween the gate driver pulses from DSP to the VSI. Experimental
results are discussed as follows.
A. MPPT Performance of the PV Array
Fig. 6(a) and (b) shows the execution of MPPT along with
PV current versus PV voltage and PV power versus PV voltage
characteristics, respectively. The small circle denotes the oper-
ating point as well as MPPT performance in percentage. It is
noticed from the figure that MPPT percentage is close to 100%.
Therefore, under rated condition as well as under varying atmo-
spheric conditions, the pumping system extracts the maximum
energy from the PV array.
B. Starting Characteristics of the Proposed System
Fig. 7 shows the variation in the parameters at starting.
Recorded waveforms show dc bus voltage Vdc , PV current
IPV , IM current ima , and IM speed Nr at 1000 W/m2 radia-
tion. Initially, the boost converter is switched OFF, due to MPPT
 3650 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 54, NO. 4, JULY/AUGUST 2018

Fig. 5. Steady-state and transient behaviors of the proposed system.

Fig. 7. Experimental characteristics of the system at starting.

algorithm being dysfunctional. After starting the IMD, the boost

converter is switched ON at 1200 rpm and the PV array starts
operating at MPP. The PV array is found to be in operation with
the efficiency of more than 99%.

C. Steady-State Characteristics of the Proposed System

Fig. 6. (a) Performance of the system at 1000 W/m2 radiation. (b) Performance
of the system at 500 W/m2 radiation.
Fig. 8 shows the recorded waveforms of the system the under
steady-state condition. The dc bus voltage Vdc , PV current IPV ,
motor current ima , and motor speed Nr are shown in the figure.
Fig. 9 shows Vdc , D, IPV , and Nr . Under the steady-
state condition, the duty ratio is found to be 0.12. At this

Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
SINGH et al.: STANDALONE PHOTOVOLTAIC WATER PUMPING SYSTEM USING INDUCTION MOTOR DRIVE WITH REDUCED SENSORS 3651

Fig. 8. Steady-state response of the system V dc , IPV , ima , and N r .

Fig. 11. Steady-state characteristics of the system V SW , iL , and V D .

Fig. 9. Steady-state characteristics of the system V dc , D, IPV , and N r .

Fig. 12. Steady-state characteristics of the system T e , D, ima , and N r .

steady-state condition, the motor operates at 1330 rpm. The

purpose of Fig. 9 is to show the dynamic variation of the duty
ratio in steady state. There are slight variations in the duty
ratio due to the real-time estimation of the MPP. Vdc , ima , imb ,
and imc are shown in Fig. 10. Moreover, Fig. 11 shows voltage
across switch Vsw , current through the inductor iL , and the
voltage across diode VD . The time at which the IGBT is ON, the
voltage across the IGBT is zero and a small fraction of voltage
remains across the diode in this duration. Moreover, during this
period, the inductor stores the energy, and in the subsequent
period, it releases the energy to the dc bus.
In Fig. 12, the torque generated Te , duty ratio of the boost
converter D, motor current ima , and motor speed Nr are shown.
Fig. 12 shows the mechanical parameters (torque, speed, and
power) in single scope and the fine perturbation in the duty
Fig. 10. Steady-state characteristics of the system V dc , ima , imb , and imc .
ratio.

Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
 3652
TABLE II
COMPARISON OF THE COPPER LOSSES IN TWO DIFFERENT TOPOLOGIES
The bold text represents the data of the proposed topology (Two Stage System).
Fig. 13. Response under decrease in radiation from 1000 to 500 W/m2 .
Fig. 14. Dynamic characteristics under decrease in radiation from 1000 to
500 W/m2 .
D. Dynamic Performance of the Proposed System
The variation in the incident radiation is tested using the
PV simulator to show the satisfactory MPPT tracking. Initially,
the radiation was 1000 W/m2 , which is reduced to 500 W/m2 .
Fig. 13 shows a variation in Vdc , IPV , im , and Nr . It can be
Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 54, NO. 4, JULY/AUGUST 2018
deduced from the recorded waveforms that the dc link voltage
is maintained constant and the PV current is reduced to half.
Fig. 14 shows the variation in the above-mentioned parameters
with an increase in the radiation from 500 to 1000 W/m2 .
VI. COMPARISON WITH SINGLE-STAGE TOPOLOGY
The two-stage system with dc bus clamping is studied along
with the single-stage system with variable dc bus. The dc bus
in the single state system is variable, since it is tied to the PV
array; hence, with the variation in solar insolation, the PV array
voltage reduces. A comparison of losses for single-stage and
two-stage systems is given in Table II. It consists of developed
ac voltage calculation at the terminals of the motor, and the
line current is calculated by considering the power factor of the
induction motor to be 0.8. Copper losses of an induction motor
are calculated in both topologies with the calculated current.
It is observed that the copper losses are always higher in the
single-stage system. It can be inferred from the calculation that
as the radiation decreases, the difference in the copper losses
decreases because of the reduction in the current. Topology I
refers to a single-stage system, while Topology II refers to a
two-stage system. Since the rated efficiency of the motor is 81%
at rated voltage, the constant losses are calculated. These are
considered to be constant throughout the operation
Pout Pout
η = 0.81 = =
Pin 2200
Pout = 1782 W
Pcore loss = Pin − Pout − Pcu
Pcore loss = 2200 − 1782 − 3 × 8.32 ×(0.603 + 0.7) = 148.7 W.
The motor current is calculated from the input power (peak
power available from PV array under any condition) as
√
Pin = 3 × Vll × Is × pf.
VII. MAIN CONTRIBUTION OF THE PROPOSED
CONTROL SCHEME
The proposed control system has a salient feature of being
immune to the variation in the estimation of the pump constant.
Moreover, the frictional loss across the pump column is well
SINGH et al.: STANDALONE PHOTOVOLTAIC WATER PUMPING SYSTEM USING INDUCTION MOTOR DRIVE WITH REDUCED SENSORS 3653

Fig. 15. Influence of the wrong estimation of the pump constant.

taken care off by the proposed control. A base speed/frequency
reference is estimated from the MPPT algorithm, which depends
on the pump constant Kpump . However, an additional term is sub-
tracted from this base speed/frequency, which is obtained from
the PI controller. The error in dc bus voltage corresponds to the
imbalance in the active power in the system and the losses of
the converter. In the absence of the feedforward term, the esti-
mated reference speed is generated by the PI controller. Hence,
the performance is sluggish and dynamic behavior of the sys-
tem is also not satisfactory. Moreover, even if a wrong value
of the pump constant is chosen, the proposed control system
estimates the reference speed accurately. Fig. 15 shows the per-
formance of the proposed system with two pump constants. One
of these constants deviates from the actual value. In the figure,
the blue line corresponds to the control with the actual value
of the pump constant, i.e., 6.554 × 10−4 N · m/(rad/s)2 , while
the red curve depicts the performance when the pump constant
is 8.025 × 10−4 N · m/(rad/s)2 . It is interesting to note that the
output of the PI controller is about 4 rpm in the blue curve,
while that in the red curve is −88 rpm. The feedforward term or
the base speed is 1372 rpm in the blue curve, while it reduces
to 1286 rpm in the red curve. This is because of an increase in
the fed value of the pump constant. However, in both the cases,
the subtraction of these quantities gives the accurate reference
speed for the extraction of the maximum power from the PV
array. The value of reference frequency is 45.8 Hz in both the
cases. Moreover, the dc bus voltage is settled to a reference value
of 400 V. The proposed control algorithm is inherently immune
to the pump constant.
Fig. 16(a) and (b) shows the overall cost effectiveness of the
proposed system. The cost estimation is made from [24] and
Fig. 16. (a) Brief cost estimation of the proposed solar water pumping system.
(b) Cost bifurcation of the overall system.
[25] in Indian rupees (INR). The analysis includes the bulk
cost of major components. The total cost considering the major
components is estimated to be INR 6087, as shown in Fig. 16(a).
Moreover, a pie chart illustrating the contribution of each com-
ponent is shown in Fig. 16(b).
VIII. CONCLUSION
A standalone PV water pumping system with a reduced sen-
sor has been proposed. It utilizes only three sensors eliminating
position and line current sensors. The reference speed gener-
ation for the V/f control scheme has been proposed based on
the available power by regulating the active power at dc bus.
The PWM frequency and pump affinity law have been used to
control the speed of an induction motor drive. Its feasibility of
operation has been verified through simulation and experimen-
tal validation. Various performance conditions such as starting,
variation in radiation, and steady state have been experimentally
verified and found to be satisfactory. The main contribution of
the proposed control scheme is that it is inherently immune to
the error in estimation of the pump constant. The system tracks
the MPP with acceptable tolerance even at varying radiations.
APPENDIX
Parameters of the proposed system: 2.2 kW (3 HP), 230 V,
8.2 A, 50 Hz three phase, 1430 rpm, 4 pole, Rs = 0.603 Ω,
Rr = 0.7 Ω, Xs = 1.007 Ω, Xr = 0.9212 Ω, and Xm =
23.56 Ω.

Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
 3654 IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS, VOL. 54, NO. 4, JULY/AUGUST 2018
REFERENCES
[1] E. Drury, T. Jenkin, D. Jordan, and R. Margolis, “Photovoltaic investment
risk and uncertainty for residential customers,” IEEE J. Photovolt., vol. 4,
no. 1, pp. 278–284, Jan. 2014.
[2] E. Muljadi, “PV water pumping with a peak-power tracker using a simple
six-step square-wave inverter,” IEEE Trans. Ind. Appl., vol. 33, no. 3,
pp. 714–721, May/Jun. 1997.
[3] U. Sharma, S. Kumar, and B. Singh, “Solar array fed water pump-
ing system using induction motor drive,” in Proc. 1st IEEE Intern.
Conf. Power Electron., Intell. Control Energy Syst., Delhi, India, 2016,
pp. 1–6.
[4] T. Franklin, J. Cerqueira, and E. de Santana, “Fuzzy and PI con-
trollers in pumping water system using photovoltaic electric genera-
tion,” IEEE Trans. Latin Amer., vol. 12, no. 6, pp. 1049–1054, Sep.
2014.
[5] R. Kumar and B. Singh, “BLDC motor-driven solar PV array-fed wa-
ter pumping system employing zeta converter,” IEEE Trans. Ind. Appl.,
vol. 52, no. 3, pp. 2315–2322, May/Jun. 2016.
[6] S. Jain, A. K. Thopukara, R. Karampuri, and V. T. Somasekhar, “A single-
stage photovoltaic system for a dual-inverter-fed open-end winding induc-
tion motor drive for pumping applications,” IEEE Trans. Power Electron.,
vol. 30, no. 9, pp. 4809–4818, Sep. 2015.
[7] J. Caracas, G. Farias, L. Teixeira, and L. Ribeiro, “Implementation of a
high-efficiency, high-lifetime, and low-cost converter for an autonomous
photovoltaic water pumping system,” IEEE Trans. Ind. Appl., vol. 50,
no. 1, pp. 631–641, Jan./Feb. 2014.
[8] R. Antonello, M. Carraro, A. Costabeber, F. Tinazzi, and M.
Zigliotto, “Energy-efficient autonomous solar water-pumping system for
permanent-magnet synchronous motors,” IEEE Trans. Ind. Electron.,
vol. 64, no. 1, pp. 43–51, Jan. 2017.
[9] M. Calavia1, J. M. Perié1, J. F. Sanz, and J. Sallán, “Comparison of MPPT
strategies for solar modules,” in Proc. Int. Conf. Renew. Energies Power
Quality, Granada, Spain, Mar. 22–25, 2010, pp. 1440–1445.
[10] Trishan Esram and P. L. Chapman, “Comparison of photovoltaic array
maximum power point tracking techniques,” IEEE Trans. Energy Con-
vers., vol. 22, no. 2, pp. 439–449, Jun. 2007.
[11] B. Subudhi and R. Pradhan, “A comparative study on maximum power
point tracking techniques for photovoltaic power systems,” IEEE Trans.
Sustain. Energy, vol. 4, no. 1, pp. 89–98, Jan. 2013.
[12] A. Garrigos, J. Blanes, J. Carrascoa, and J. Ejea, “Real time estima-
tion of photovoltaic modules characteristics and its application to max-
imum power point operation,” Renew. Energy, vol. 32, pp. 1059–1076,
2007.
[13] B. Singh, S. Kumar, and C. Jain, “Damped-SOGI-based control algorithm
for solar PV power generating system,” IEEE Trans. Ind. Appl., vol. 53,
no. 3, pp. 1780–1788, May/Jun. 2017.
[14] X. D. Sun, K. H. Koh, B. G. Yu, and M. Matsui, “Fuzzy-logic based V/f
control of an induction motor for a DC grid power leveling system using
flywheel energy storage equipment,” IEEE Trans. Ind. Electron., vol. 56,
no. 8, pp. 3161–3168, Aug. 2009.
[15] S. R. Bhat, A. Pittet, and B. S. Sonde, “Performance optimization of
induction motor-pump system using photovoltaic energy source,” IEEE
Trans. Ind. Appl., vol. IA-23, no. 6, pp. 995–1000, Nov. 1987.
[16] Y. Yao, P. Bustamante, and R. Ramshaw, “Improvement of induction mo-
tor drive systems supplied by photovoltaic arrays with frequency con-
trol,” IEEE Trans. Energy Convers., vol. 9, no. 2, pp. 256–262, Jun.
1994.
[17] U. Sharma, B. Singh, and S. Kumar, “Intelligent grid interfaced solar
water pumping system,” IET Renewable Power Gener., vol. 11, no. 5,
pp. 614–624, Mar. 2017.
[18] F. Karbakhsh, M. Amiri and H. A. Zarchi, “Two-switch flyback inverter
employing a current sensorless MPPT and scalar control for low cost solar
powered pumps,” IET Renewable Power Gener., vol. 11, no. 5, pp. 669–
677, 2017
[19] S. Jain, R. Karampuri, and V. Somasekhar, “An integrated control algo-
rithm for a single-stage PV pumping system using an open-end winding
induction motor,” IEEE Trans. Ind. Electron., vol. 63, no. 2, pp. 956–965,
Feb. 2016.
Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
[20] A. Achour, D. Rekioua, A. Mohammedi, Z. Mokrani, T. Rekioua, and S.
Bacha, “Application of direct torque control to a photovoltaic pumping
system with sliding-mode control optimization,” Elect. Power Compon.
Syst., vol. 44, no. 2, pp. 172–184, 2016.
[21] C. Slabbert and M. Malengret, “Grid connected/solar water pump for rural
areas,” in Proc. IEEE Int. Symp. Ind. Electron., Pretoria, South Africa,
1998, vol. 1, pp. 31–34.
[22] B. Singh, A. Chandra, and K. Al-Haddad, Power Quality: Problems and
Mitigation Techniques. Chichester, U.K.: Wiley, 2015.
[23] N. Mohan, T. Undeland, and W. Robbins, “Power Electronics: Converters,
Applications and Design. New Delhi, India: Wiley, 2009, vol. 3.
[24] [Online]. Available: http://in.rsdelivers.com
[25] [Online]. Available: http://www.mouser.in
Bhim Singh (SM’99–F’10) was born in Rahamapur,
Bijnor, Uttar Pradesh, India, in 1956. He received the
B.E. degree in electrical engineering from the Univer-
sity of Roorkee [now Indian Institute of Technology
(IIT) Roorkee], Roorkee, Uttarakhand, India, in 1977
and the M.Tech. degree in power apparatus and sys-
tems and the Ph.D. degree both from IIT Delhi, New
Delhi, India, in 1979 and 1983, respectively.
In 1983, he joined as a Lecturer the Department
of Electrical Engineering, IIT Roorkee, where he be-
came a Reader in 1988. In December 1990, he joined
as an Assistant Professor the Department of Electrical Engineering, IIT Delhi,
India, where was promoted to an Associate Professor in 1994 and a Professor
in 1997. From September 2007 to September 2012, he was an ABB Chair Pro-
fessor. From October 2012 to September 2017, he was a CEA Chair Professor.
From July 2014 to August 2016, he was the Head of the Department of Electrical
Engineering, IIT Delhi. Since August 2016, he has been the Dean, Academics at
IIT Delhi. He has guided 69 Ph.D. dissertations, and 167 M.E./M.Tech./M.S.(R)
theses. He has filed 31 patents. He has executed more than 80 sponsored and con-
sultancy projects. He has coauthored a text book on power quality Power Quality
Problems and Mitigation Techniques (Wiley, 2015). His research interests in-
clude solar PV grid interface systems, microgrids, power quality monitoring
and mitigation, solar PV water pumping systems, improved power quality ac–
dc converters, power electronics, electrical machines, drives, flexible alternating
transmission systems, and high voltage direct current systems.
Dr. Singh was the recipient of the Khosla Research Prize of the University
of Roorkee in 1991, the JC Bose and Bimal K Bose Awards of The Institution
of Electronics and Telecommunication Engineers (IETE) for his contribution in
the field of power electronics, the Maharashtra State National Award of Indian
Society for Technical Education for his outstanding research work in the area
of power quality, the PES Delhi Chapter Outstanding Engineer Award in 2006,
the Khosla National Research Award of IIT Roorkee in 2013, the Shri Om
Prakash Bhasin Award in 2014 in the field of engineering including energy and
aerospace, the IEEE PES Nari Hingorani Custom Power Award in 2017, and
the “Faculty Research Award as a Most Outstanding Researcher” in the field
of engineering in 2018 by Careers360, India. He has been the General Chair
of the 2006 IEEE International Conference on Power Electronics, Drives and
Energy Systems, the General Co-chair of the 2010 IEEE International Confer-
ence on Power Electronics, Drives and Energy Systems, the General Co-chair
of the 2015 IEEE International Conference, the General Co-chair of 2016 IEEE
International Conference held in New Delhi, and the General Co-Chair of 2017
National Power Electronics Conference held in Pune. He is the fellow of the In-
dian National Academy of Engineering, the Indian National Science Academy,
The National Academy of Science, India, the Indian Academy of Sciences, The
World Academy of Sciences, the Institute of Engineering and Technology, The
Institution of Engineers (India), and The Institution of Electronics and Telecom-
munication Engineers and a Life Member of the Indian Society for Technical
Education, System Society of India (SSI), and the National Institution of Quality
and Reliability. He is the JC Bose fellow of DST, Government of India since
December 2015.
SINGH et al.: STANDALONE PHOTOVOLTAIC WATER PUMPING SYSTEM USING INDUCTION MOTOR DRIVE WITH REDUCED SENSORS
Utkarsh Sharma (M’17) was born in Kota,
Rajasthan, India, in 1991. He received the B.Tech.
degree in electrical engineering from Sardar
Vallabhbhai National Institute of Technology, Surat,
Gujarat, India and the M.Tech. degree in power
electronics, electrical machines and drives from the
Indian Institute of Technology Delhi (IIT Delhi), New
Delhi, India, in 2013 and 2016, respectively. He is
currently working toward the Ph.D. degree at the De-
partment of Electrical Engineering, IIT Delhi, New
Delhi.
His research interests include power electronics, control of electrical drives,
renewable energy applications and design of special electrical machines.
Mr. Sharma was the recipient of the National Talent Search Scholarship
awarded by the National Council of Educational Research and Training, New
Delhi, India. He has received the Best Industry Relevant M.Tech. Thesis Award
from the Foundation for Innovation and Technology Transfer by IIT Delhi in
2016.
Authorized licensed use limited to: GOVERNMENT COLLEGE OF ENGINEERING. Downloaded on May 18,2022 at 17:13:20 UTC from IEEE Xplore. Restrictions apply.
3655
Shailendra Kumar (S’15–M’17) was born in
Mahoba, Uttar Pradesh, India, in 1988. He received
the B.Tech. degree in electrical and electronics en-
gineering from Uttar Pradesh Technical University,
Lucknow, Uttar Pradesh, India and the M.Tech. de-
gree in power electronics, electrical machine and
drives from the Indian Institute of Technology Delhi
(IIT Delhi), New Delhi, India, in 2010 and 2015, re-
spectively. He is currently working toward the Ph.D.
degree at the Department of Electrical Engineering,
IIT Delhi, New Delhi.
His research interests include power electronics, power quality, custom power
devices, and renewable energy.
Mr. Kumar was the recipient of the POSOCO Power System Award from the
Foundation for Innovation and Technology Transfer by IIT Delhi in 2016 and
the IEEE UPCON Best Paper Award in 2016.